HIGH SCHOOL STUDENTS’ HEALTHY EATING BEHAVIOR PREDICTED BY THEIR MOTIVATION TO PARTICIPATE IN URBAN GARDENING ACTIVITIES

Urban high school students often face significant barriers to healthy eating, including limited nutritional knowledge, lack of access to nutritious foods, and social pressures promoting unhealthy diets. This research aimed to describe and predict urban high school students’ gardening motivation and healthy eating behavior after participating in urban gardening activities, employing the Situated Expectancy-Value Theory (SEVT) framework. A quantitative research design was used, involving 113 participants from two urban high schools who completed a questionnaire measuring intrinsic value, cost value, utility value (personal and local context), self-efficacy (cultural and gardening), health motivation, and healthy eating behavior. An exploratory factor analysis was performed to validate the measurement instrument, followed by multiple linear regression to predict students' eating behavior based on their motivational constructs after engaging in gardening activities. Conclusions indicated that the questionnaire effectively measured five SEVT motivation constructs: personal and community usefulness, health motivation, intrinsic value, cultural gardening self-efficacy, and cost value. Students who participated in urban gardening activities reported healthier eating behaviors than before the participation and compared to their peers who did not participate in any activity. Students found the gardening activities useful and motivating, especially at the personal and community level, with moderate intrinsic and cultural motivation and low perceived cost. Urban gardening activities and three motivational factors explained 30% of the variance in students’ healthy eating behaviors, with personal and community usefulness and intrinsic value being the strongest predictors. Recommendations for future research include exploring additional factors not considered in this study that may increase the explained variance of the predictive model.